Inorg. Chem. 1991, 30, 2949-2952 in Figure 6B. The presence of both couples in the film is evidenced by the waves that appear at +0.88 and +0.70 V corresponding to the RulI1/I1and FelI1/I1couples, respectively. By integration, the ratio of Fe2+ to Ru2+ in the copolymeric film was 16:84. Copolymerization of [R~(bpy)~(4,4'-(CH~Br)~bpy)]~+ and [Os(vbpy),12+. A solution 1.08 M in [R~(bpy)~(4,4'(CH2Br)2bpy)2+and 0.73 M in [0s(vbpy),12+ (6040 RulI:OsI1) was electropolymerized by scanning the potential to -1.4 V. At this potential only the Ru2+complex was reduced. This experiment was conducted to see if copolymerization could be induced by selective reduction at the Ru2+ complex. In the cyclic voltammogram of the resulting film reversible waves appeared at +0.41 and +1.00 V for the Oslll/ll and Rulll/ll couples, respectively. By integration, the ratio of Ru" to Ost1was 73:27. In a control experiment in which there was only [Os(vbpy),I2+ in the polymerization solution, no significant electropolymerization was observed. Layered Structures. In order to facilitate this part of the presentation, layered structures will be represented in the format: electrode/component 1/component 2. In this representation component 1 is the initially electropolymerized monomer. After electropolymerization, it is at the electrode/film interface. Component 2 is the monomer that was electropolymerized as a second film,atop the first, in a second step. After the second electropolymerization it is at the film/solution interface. R / [ F ~ ( ~ ~ ' - ( C W W P),P+/[Ru(bpy)2(44'-(~2Br,2bpy>P'. Y The preparation of an inner layer of electropolymerized [Fe(4,4'-(CH2Br)2bpy)3]2+(I' = 5.4 x lo4 mol/cm2) was carried out initially. This step was followed by the electropolymerization in a second step. A cyclic of [R~(bpy)~(4,4'-(CH~Br)~bpy)]~+ voltammogram for the resulting "bilayer" is shown in Figure 7A. In the voltammogram, waves for both the RulI1/I1and FelI1/** couples were observed. The reductive component for the RulI1/I1 wave appeared as a shar prepeak at the onset of the reductive component for the FelI1/Irwave. This is due to an enhancement in the rate of reduction of the Ru3+ in the outer layer. At the onset of reduction of Fe3+in the inner layer, the Fe2+produced
2949
can, in turn, reduce Ru3+,which is spatially isolated in the outer layer. Thus,reduction of Ru3+is mediated by F$+. This behavior is in contrast to the copolymeric film in Figure 6A where neither of the component waves is distorted. Both components are in the inner layer and have direct channels for undergoing electron transfer with the electrode surface. Pt/[Fe(4,4'-(CHzBr)zbpy)3]2+/[Ru(vbpy)3~2+. An inner layer of electropolymerized [Fe(4,4'-(CH2Br)2bpy)3]z+(I' = 7.8 X 1 P mol/cmz) was prepared on a Pt electrode. In a second electropolymerization step an outer layer of polymerized [Ru(vbpy),12+ was added. A cyclic voltammogram of the resulting film is shown in Figure 78. The overlap between the RulI1/I1and Fenl/I1waves is greater than in the previous case. A sharp prepeak is still observed at the onset of the FelI1/I1reduction due to the mediated reduction of Ru3+ in the outer layer by Fez+ in the inner layer. Pt/[Ru(vbpy ),I2+/[Fe( 4,4'-(CH2Br)2bpy)3]2+. A layer of polymerized [Ru(vbpy),12+ (I' = 7.6 X 10-9 mol/cmz) was formed by reductive electropolymerization. In a second step a layer of was added. In electropolymerized [ Fe(4,4'-(CH2Br)2bpy)3]2+ Figure 7C is shown a cyclic voltammogram of the resulting bilayer film. The sharp prepeak observed at the onset of the oxidative component for the RulI1/" couple is a consequence of the bilayer structure. The outer layer, which contains Fe2+,is isolated from the electrode. Oxidation of Fez+requires mediation by the R 3 + sites in the inner layer. Reductive components for both the RUI"/~I and FelI1/" couples are observed on the return scan. The reduction of Fe3+to Fe2+occurs because of the large overlap between the RulI1/I1and FelI1/I1waves.18a Fe3+ in the outer layer formed in the oxidative scan cannot be trapped. Acknowledgment is made to the Army Research Office under Grant No. DAAL03-90-G-0062 for support of this work and to Kimberly Hooker for providing the XPS results. XPS instrumentation was funded by an instrumentation grant (R. W. Linton, Principal Investigator) from the Office of Naval Research, ONR NO. N00014-86-G-0200.
Notes Contribution from the Science and Engineering Research Laboratory, Waseda University, Ohkubo 3-4-1, Shinjuku-ku, Tokyo 169, Japan Photochemistry of Chlorochromium(II1) Tetraphenylporphyrinatein Acetone. Studies on Photodissociation of the Axial Acetone in the Temperature Range 180-295 K by Laser Photolysis Minoru Yamaji'
Received August 15, 1990 It is well-known that natural and synthetic metalloporphyrins undergo photodissociation of the axial ligand such as CO, NO, 02, CN-, or pyridine upon light irradiation. By means of picoand nanosecond laser photolyses, many studies of the reactive state for the ligand dissociation as well as on deactivation processes of excited states have been carried out.' These studies help extensively to understand the effects of the axial ligand on the excited-state behavior of metalloporphyrins. For chromium porphyrins, in our previous pa r,2 it was reported that the pyridinate of C1Cr1I1(TPP),C1CrqTPP)Py (Py = pyridine), in acetone photodissociates the axial pyridine to Prucnt address: Department of Chemistry, Faculty of Engineering, Gunma University, Kiryu, Gunma 376, Japan.
produce C1Cr1I1(TPP) with a quantum yield of 0.16 upon both 355- and 532-nm laser irradiation at room temperature. Since no excitation wavelength dependence was observed, we had three candidates for the reactive state for dissociation: the tripletsextet, 6T1,the tripletquartet, 'TI, and the singletquartet, 'SI,states.' On the basis of the following two reasons, neither the 'TI nor the 6TI state is the reactive state: (1) the quantum yields for photodissociation are independent of the concentrations of oxygen; (2) the 6TI state, which is in thermal equilibrium with the 'TI state, is quenched by oxygen. Therefore, the 45, state wsa concluded to be the reactive state for the dissociation of Py from C1Cr1I1(TPP)Py. In the present work, we have investigated the photochemistry of the photodissociation of the axial acetone from the acetonate of C1Cr1I1(TPP),C1Cr1I1(TPP)Ac(Ac = acetone), by laser pho(1) (a) Lavalette, D.; Tetreau, C.; Momenteau, M.J. Am. Chem. Soc. 1979,
101,5395-5401. (b) Hoshino, M.; Kashiwagi. Y. J . Phys. Chem. 1990, 94,673478. (c) Hoshino, M. Chem. Phys. Lerr. 1985,120,5&52. (d) Hoffman, B. M.; Gibson, Q. H. Proc. Narl. Acad. Sci. U S A . 1978,75, 21-25. (e) Hoshino, M.; Arai. S.; Yamaji. M.; Hama, Y. J . Phys. Chem. 1985, 90, 2109-21 1 1 . (f) Tait, C. D.; Hoiten, D.; Gouterman, M.J . Am. Chem. Soc. 1 M , 106,6653-6659. (g) Hoshino, M.; Kogure, M.;Amano, K.; Hinohara, T. J . Phys. Chem. 1989, 93, 6655-6659. (2) Yamaji, M.; Hama, Y.; Hoshino, M. Chem. Phys. Lcrr. 1990, 165, i n e iI 4... (3) G o u t c a n , M.; Hanson. L. K.; Khalil. G.-E.; Leenstra, W. R.; Buchler, J. W. J . Chem. Phys. 1975,62, 2343-2353.
--_ - _
0020- 1669191/ 1330-2949$02.50/0 Q 1991 American Chemical Society
2950 Inorganic Chemistry, Vol. 30, No. 14, 1991
Notes 0.2
3
-
0.1 r
I
Y
z 2
a 2
In
9
'9
0
X
Y
0
y - 0.1
u , 1
a
m a
0
2 - 0.2 0
400
450
WAVELENGTH , nm
a
- 0.3
Figure 1. Absorption spectra of CICrlI1(TPP)in (a) acetone and (b) CH2C12at room temperature.
Y I
tolysis in the temperature range 180-295 K. On the basis of the quantum yield measurements of the photodissociation of Ac and the 6T1state at various temperatures, the reactive state for the dissociation of Ac from CICr"'(TPP)Ac as well as deactivation processes of the excited chromium(II1) porphyrins is discussed in detail. Experimental Section CICriI1(TPP)was synthesized and purified according to the literature procedure.' Reagent grade acetone and dichloromethane were used as supplied. Absorption and luminescence spectra were recorded on a Hitachi 330 spectrophotometer and a Hitachi MPF 4 fluorescence spectrophotometer, respectively. Laser photolysis studies were carried out by using the second (532 nm) and third harmonics (355 nm) of a Nd-YAG laser from J. K. Laser Ltd.: the pulse duration and energy were 20 ns and 100 mJ/pulse, respectively. The detection system used for observation of transient spectra has been described elsewhere.5 Temperatures of sample solutions were controlled by a cryostat from Oxford Instruments (Model D N l0200) with a precision of Al.0 OC. Sample solutions were prepared in the dark to prevent photodecomposition of CICrlIi(TPP) and were degassed to ca. 1odTorr on a vacuum line by freeze-pump-thaw cycles.
Results and Discussion Figure 1 shows absorption spectra of CICrlI1(TPP)in acetone and dichloromethane at room temperature. The molar absorption coefficients were determined as 3.0 X los M-'cm-I at 446 nm and 2.6 X los M-' cm-I at 451 nm in acetone and dichloromethane, respectively. Since CH2C12is regarded as a noncoordinating solvent, the difference in the peak wavelength between these spectra indicates that, in an acetone solution, CICr"'(TPP) is coordinated with an acetone molecule, Ac, as the sixth ligand to form CICr"'(TPP)Ac. Figure 2 shows the transient spectra observed for a degassed acetone solution of CICr"'(TPP)Ac (4.02 X lod M) at (a) 50 ns and (b) 2 ps after 355-nm laser pulsing at room temperature. Transient spectrum a decays within 200 ns, leaving transient spectrum b, which returns t o the spectrum of CICr'"(TPP)Ac with the first-order rate constant kd = 5.7 X IO3 s-'. No residual absorption was observed after the decay of transient b. The fast decay component of the transient spectra obtained by subtracting transient spectrum b from transient spectrum a is ascribed to the 6TI state of CICr111(TPP)Ac.2 Transient spectrum b having the positive peak at 455 nm and the negative one at 445 nm is quite similar to the difference spectrum obtained by subtracting the spectrum of CICr"'(TPP) in CHZCIzfrom that in acetone. As described above, CH2CI2is
.
.
.
I
.
.
I
450
400
WAVELENGTH , nm
Figure 2. Transient spectra observed for an acetone solution of ClCr"'(TPP) (4.02 X lod M) after a 355-nm laser pulse at (a) 50 ns (A) and (b) 2 ps (0)at room temperature.
regarded as a noncoordinating solvent. Thus, transient b is ascribed to C1Cr1I'(TPP) having no axial acetone: CICrlI1(TPP)Ac C1Cr1I1(TPP)Ac ClCr"'(TPP)
hv
6T,
CICrI1'(TPP) + Ac
+ Ac -% CICr"'(TPP)Ac
Photoexcitation of CICr"'(TPP)Ac gives the 6TI state and causes dissociation of the axial Ac to yield five-coordinated CICr"'(TPP), which recombines with Ac to regenerate ClCr"'(TPP)Ac with the rate constant kd = kA,[Ac]. The value of kAcwas calculated as 4.2 X lo2 M-I s-'. This value is in a good agreement with that obtained from the studies on photodissociation of ClCr"*(TPP)Py in an acetone solution (4.2 X lo2 M-l s-I).~ The quantum yield, &is, for photodissociation of the axial acetone from ClCr"'(TPP)Ac was determined with the laser photolysis technique. The number of photons absorbed by CIat a laser excitation wavelength (532 or 355 Cr"'(TPP)Ac, lab, nm) was determined by measuring the triplet-triplet (T-T) absorption of zinc(I1) tetraphenylporphyrinate, Zn"(TPP), in benzene. The triplet yield, &, and the molar absorption coefficient, €470, at 470 nm of the triplet state of Zn"(TPP) have been reported as 0.83 and 7.3 X lo4 M-' cm-', respectively.6 After laser pulsing, the initial absorbance change, A&, observed for formation of the triplet state of Zn"(TPP) is represented as ADT = 4 ~ f 4 7 d a k (1) On the other hand, when an acetone solution of CICr"'(TPP)Ac has the same absorbance at the irradiation wavelength as that of the benzene solution of ZnI1(TPP) and is subjected t o laser irradiation, the quantum yield, qjdL, for photodissociation of Ac from ClCr"'(TPP)Ac is expressed as
(2) Here, ADdLand At are the absorbance change at 445 nm observed at 2 ps after laser pulsing and the difference in the molar absorption coefficient at 445 nm between ClCr"'(TPP)Ac in acetone and CICrlI1(TPP) in CH2C12(1.32 X lo5 M-' cm-I). With the use of eq 1 , eq 2 is transformed to bdis
4dis
(4) Summerville, D. A.; Jones, R. D.; Hoffman, B. M.;Basolo, F.J . Am. Chem. Soc. 1977, 99, 8195-8202. (5) Hoshino, M.; Imamura, M.; Watanabe, S.; Hama, Y. J . Phys. Chem. 1984, 88, 45-49.
.
=
=
mdis(Af)-'Zats-'
mdis(AD~)-'f470(Af)-'~
(3)
(6) Hurley. J. K.; Sinai, N.; Linschitz, H. J . Phorochcm. Phorobfol. 1983, 38, 9-14.
Inorganic Chemistry, Vol. 30, No. 14, 1991 2951
Notes
,
1.0 L
1.0
kOEXP(-AEd / RT)
s,
I
IkN 0
200
250
300
'
dissociation kST
0
TEMPERATURE ( K 1
Figure 3. Plot of the quantum yields, #&, for dissociation of the axial acetone ( 0 )and the yields, h, of the 6TI state (A) in the temperature range 180-295 K. The solid and broken lines are the values calculated according to the reaction scheme (see text).
The quantum yields for photodissociation of the axial acetone at 295 K were determined as 0.63 f 0.03 and 0.55 0.05 upon 355- and 532-nm laser irradiation, respectively. These results reveal that the quantum yield for photodissociation of Ac has no excitation wavelength dependence within an experimental error. In agreement with the present case of ClCr1I1(TPP)Ac,it is reported that the quantum yield for photodissociation of the axial pyridine from CICrlI1(TPP)Py is independent of the irradiation wavelength.2 By the analogy to eq 2, the quantum yield for the formation is expressed with the use of the molar abof the 6TIstate, h, sorption coefficient of the 6Tl state at 470 nm, em, and the abat 470 nm originating from the formation sorbance change, -, of the 6T1 state as follows:
I IkQ I ks 4u SO
*
h = ADmcm-'I*b-'
(4) However, because of the difficulty in determining the value of -, the value of h could not be obtained experimentally. Since the products that we observed upon irradiation of CICr1I1(TPP)Ac were merely CICrI1*(TPP)and the 6TI state at 295 K, we assumed the value of h at 295 K as follows: h ( 2 9 5 ) = 1 - 4d,(295) (5) where &(295) is the quantum yield (=0.63) for photodissociation at 295 K. From eq 5, we obtain t # ~ ~ ( 2 9 5=)0.37. In order to study the temperature dependence off$& and b, laser photolysis studies were executed with a cryostat in the temperature range 180-295 K. According to eqs 2 and 4, the values of 4di and 4m at lower temperatures than 295 K can be determined by measurement of the absorbance changes, fi&(T) and AD,(T), at a given temT), perature, T,respectively: the quantum yields, 4&( 7) and b( at a given temperature, T,are respectively represented as #dis(T) = Udii(T) udi1(295)-' '$di1(295)
(6)
h(T) = &(T) A&(295)-1 4Ts(295) (7) where ADdk(295)and &(295) are the absorbance changes at 295 K. By using 4di,(295) (10.63) and h ( 2 9 5 ) (=0.37), eqs 6 and 7, we obtained the values of f#db and 4m,respectively, in the temperature range 180-295 K. In Figure 3, the obtained values of (Pdir and h are represented in the temperature range studied. It was found that, from high to low temperatures, the quantum yield, f$dh, decreases while h increases. The increase in the values of at higher temperatures implies that the activation energy, A&, is necessary for dissociation of the axial Ac. In the previous study,*it was suggested that photodissociation of the axial pyridine from CICr"'(TPP)Py occurs from the 4S, state. Here, similarly, we assume that photodissociation of Ac from CICrlI1(TPP)Ac takes place via the 4S1state. In order to explain the mechanism of the dissociative reaction, we have depicted an energy diagram and decay rate processes of CICrlI1(TPP)Ac in the excited states in Figure 4. After excitation of ClCr"'(TPP)Ac by laser irradiation, the 'SI state is produced within a duration of laser pulse. Besides the process for dissociation of Ac, the 4SIstate deactivates to the
C I Cr(lll) TPP-Ac
Figure 4. Energy scheme for the ground and the excited states of CICrlI*(TPP)Ac(see text).
ground state, 'So,and the TIstates with the decay rate constants kN and ksT, respectively. Since no emission from the 4S1state is detected, kNis considered to be a rate constant for nonradiative decay. At 295 K,the assumption of eq 5 implies kN = 0. Below the 'SIstate in energy, there are two excited states, 'TIand 6T1. The energy level of the 6TI state is lower than that of the 'TIstate. Between the 4T1and the 6Tlstates, Boltzmann distribution is assumed to be established with the energy gap, AEWs3The 4TI and the 6T1state decay to the 4S0state with the rate constants kQ and ks, respectively. On the assumption that the activation energy, hEd, is necessary for dissociation, the rate constant kdir for dissociation of Ac via the 4S1state at temperature T can be described by an Arrhenius expression with the frequency factor, ko: kdir = ko exp(-AEd/RT) (8) According to the energy scheme described above, the yield for dissociation, f$dis, is formulated as follows: 4dis
= kdir(kN + kST
+ kdir)-'
(9)
By use of eq 8, eq 9 can be rewritten as 4dii
= [1 + K exP(Ud/RT)I-'
(10)
where
K = (kN + kST)/kO Equation 10 is transformed to In (4dis-l - 1) = In K
+ A&
(1 1) R'T1
(12) for easier analysis. Plots of In (4cb-' - 1) vs T I give a straight line. From the slope and the intercept of the line, we obtain A&
2.6 kcal/mol
(13)
+
K (kN ksT)/ko = 4.7 x (14) The solid line in Figure 3 plots the values of &, calculated with eqs 10, 13, and 14. Good agreement between the experimental values of 4&( T, and the solid line justifies the mechanism proposed for the dissociative process depicted in Figure 4. With a decrease in temperature, the values of tend to increase while those of +dm decrease. Concerning the values of h at lower temperatures, we assume that eq 5 holds not only at 295 K but also in the whole temperature range studied. This implies kN = 0 independent of temperature. By use of eqs 5 and 10, 4m is represented as 1 - [I + K eXp(AE,+/Rr)]-' (15) The broken line in Figure 3 represents the values of h calculated from eq 15 and the values of u d and K shown in eqs 13 and 14, &y~
Inorg. Chem. 1991, 30, 2952-2954
2952
respectively. Because the experimental values of h( 7') are in good accord with those calculated, we consider that kN = 0 holds in the temperature range studied, as assumed above. It, therefore, is likely that the photoexcited CICr"I(TPP)Ac is fated to undergo the dissociation of the axial acetone or result in the formation of the 'TI state. When the 'TIstate is formed, the 6T1state is populated, leading to establishment of Boltzmann equilibrium between the 'TI and 'TIstates.) As will be mentioned later, the excited state detected by laser photolysis of CICrlI1(TPP)Ac is principally ascribed to the 6Tl state. At room temperature, the lifetime of the 6TI state is too short, less than 100 ns, to determine precisely the decay rate constant by our laser system. However, laser photolysis at lower temperatures enables us to observe the decay profile of the transient spectrum of the 6TI state. In the temperature range 180-230 K, we could measure the decay rate constants of the 6TI state, koW, represented as3 kobd
(kS
+ xkQ)(l + x)-'
(16)
where
x = 4; exp(-UQS/RT) On the assumption x
